ApPuaE MicRoBioLoGy, Nov. 1974, p. 785-792Copyright i 1974 American Society for Microbiology

Vol. 28, No. 5Printed in U.S.A.

Effects of Seawater Cations and Temperature on ManganeseDioxide-Reductase Activity in a Marine Bacillus

W. C. GHIORSE AND H. L. EHRLICH

Department of Biology, Rensselaer Polytechnic Institute, Troy, New York 12181

Received for publication 18 July 1974

The seawater cations, Na+, K+, Mg2+, and Ca2+, each stimulated MnO2-reductase activity of whole cells and cell extracts of Bacillus 29. Concentrationsof Na+ and K+ which stimulated whole cells and cell extracts maximally wereequivalent to those in two- to fivefold diluted seawater. Cell-extract activity wasstrongly stimulated by Ca2+ and Mg2+ up to a concentration of 0.01 M Mg2+ and0.002 M Ca2+, with little additional stimulation above these concentrations.Whole-cell activity was stimulated biphasically with increasing concentrations ofCa2+ and Mg2+. Comparison of the effects of individual cations or mixtures ofthem at concentrations equivalent to their concentration in fivefold dilutedseawater showed that more activity was obtained with 0.01 M Mg2+ or 0.002 MCa2+ than with 0.1 M Na+, and more with 0.1 M Na+ than with 0.0022 M K+.Fivefold diluted seawater permitted as much or more activity as solutions ofindividual or synthetic mixtures of the cations. Pre-exposure experiments showedthat the ionic history of whole cells was important to their ultimate activity. TheMnO2-reductase activity of induced whole cells exhibited a temperature op-timum near 40 C. Cell extracts had different temperature optima (Topt),depending on whether induced glucose-linked activity (Topt = 25 C), uninducedglucose-linked, ferricyanide-dependent activity (Topt = 30 C), or uninducedferrocyanide-linked activity (Topt = 40 C) were being measured. Some of theseoptima are higher than previously reported.

The participation of bacteria in a 'marinemanganese cycle was established by studyingthe bacteriology of marine manganese nodules(2-5; 8, 15, 16). Manganese nodules yieldedbacteria which participate in both oxidationand reduction of manganese (2, 4, 6). Theenzymatic nature of these reactions has beendemonstrated (4, 5, 16). The biochemistry of theenzymatic interactions with manganese wasalso investigated in the bacteria isolated fromnodules, and characterization of the enzymesinvolved has begun (5; 8; 16; Ghiorse andEhrlich, Annu. Meet., Amer. Soc. Microbiol.,Philadelphia, Pa., Abstr. 163, p. 163, 1972).

Since bacterial reduction of Mn(IV) in nod-ules on the ocean floor must occur in a salineenvironment, it is important to determine whatinfluence, if any, the major cations of seawater(Na+, K+, Mg2+, and Ca2+) have on the process.Similarly, since bacterial Mn(IV) reduction onthe ocean floor must occur at a low temperature,it is of interest to determine what effect varioustemperatures have on this process. This reportdeals with these effects on MnO2 reduction by

I Present address: Department of Agronomy, Cornell Uni-versity, Ithaca, N. Y. 14850.

Bacillus 29, which was isolated from a man-ganese nodule from Blake Plateau in the Atlan-tic Ocean and shown to produce an inducibleMnO2-reductase system (15, 16).

(The data in this paper were taken from athesis submitted by W. C. Ghiorse to theDepartment of Biology of Rensselaer Polytech-nic Institute in partial fulfillment of the re-quirements for the Ph.D. degree).

MATERIALS AND METHODSBacterial cultures. Bacillus 29 cultures (2) were

maintained at 25 C on Stock Culture agar (Difco). Forexperimentation, uninduced cells (i.e., cells whichrequired the electron carrier, ferricyanide, for MnO2-reduction) were grown on seawater nutrient-agar inRoux slants as described previously (16). Inducedcells (i.e., those which did not require ferricyanide forMnO.-reduction) were obtained by growth in thepresence of Mn2+. This was accomplished by adding 1ml of sterile 0.1 m MnSO4-H20 to 9 ml of inoculumsuspended in seawater and distributing the mixtureover the seawater nutrient-agar surface. These cul-tures were incubated aerobically at 25 C for 24 h.

Preparation ofwhole cell suspension. Whole cellsof Bacillus 29 were removed from 1 to 12 Roux slantswith seawater. The resultant suspension was cen-

785

GHIORSE AND EHRLICH

trifuged at 10,000 x g for 10 min. Uninduced cellswere then washed two or three times in distilled waterand 1-ml samples of the final distilled-water suspen-sion was then used to inoculate the various reactionmixtures. Induced cells were washed in distilled wateror in fivefold diluted seawater, depending on theexperiment, until 9.0 ml of the wash solution was freeof Mn2+ as indicated by persulfate oxidation. Washedwhole cells were then suspended in an appropriatemedium and either assayed directly or ruptured bysonication to produce cell-free extracts.

Preparations of cell-free extracts. Cells of Bacil-lus 29 which had been washed three times in distilledwater were suspended in 5 to 15 ml of appropriateseawater or salts solution to a final cell concentrationof 2 to 4% (dry cell weight per suspension volume) andsonicated for 4 to 6 min according to the method ofTrimble and Ehrlich (16). Sonic extracts were cen-trifuged at 15,000 x g for 15 min at 4 C. Supernatantscontaining the enzymatic activity were free of celldebris when examined by phase contrast microscopy.

Assays for MnO2-reductase activity. Two assaysfor MnO,-reductase activity were used. Both weremodifications of the short-term assay of Trimble andEhrlich (15). In one modification, the glucose-linkedactivity of either ferricyanide-dependent or ferricya-nide-independent preparations was measured in du-plicate 50-ml Erlenmeyer flasks containing 0.2 g ofMnO2 (prepared as described by Trimble and Ehr-lich, 15), 5.0 ml of appropriate seawater or saltssolution, 1.0 ml of 1% glucose dissolved in appropriateseawater or salts solution, and 1.0 ml of eitherwhole-cell suspension or cell-free extract. In the caseof uninduced cell preparations, 0.1 ml of 0.01 MK,Fe(CN). was also added to the flasks. In the secondassay modification, the ferrocyanide-linked activity ofuninduced preparations was measured by altering theglucose-linked, ferricyanide-dependent assay in thefollowing ways: (i) 1.0 ml of appropriate seawater orsalts solution replaced the 1% glucose solution; (ii) 0.1ml of 0.04 M KFe(CN)..6HO or Naye(CN)..10HO replaced the 0.01 M KyFe(CN),. Both assaymixtures were incubated for 3 h. After incubation, thereaction mixtures were acidified by adding 0.05 ml of10 N HSO4 to each and incubating for 10 min at roomtemperature. The liquid portion in each reactionvessel was then centrifuged at 4,000 x g for 10 min toremove any suspended MnO,. The amounts of man-ganese in the supernatants were determined by thepersulfate oxidation method of Ehrlich (2), exceptthat 1.0-ml portions of the samples were added totubes which contained 8.0 ml of distilled water and1.0 ml of "special reagent". In all experiments suita-ble controls were assayed along with experimentalreaction mixtures. The difference in the amount ofmanganese released with and without cells or cellextract was taken as a measure of the amount ofmanganese released enzymatically. This differencewas used to compute the specific activity of wholecells and cell extracts expressed as nanomoles of Mn2+released per hour per milligram of protein. Theactivity was linear over the 3-h incubation period.

Estimation of protein content. Protein content of

cell extracts was estimated by the method of Lowry etal. (11), using bovine serum albumin as a standard.For whole cells of Bacillus 29, the average percentageof protein per cell (dry weight) was found to be 562%, based on more than 30 determinations. This valuewas used to estimate the amount of protein added toreaction flasks when specific whole-cell activity wasmeasured.

Determinations of the effects of the major cat-ions of seawater on MnO2-reductase activity. Cellsand cell-free extracts were tested in reaction mixturescontaining one or more of the chloride salts of Na, K,Mg, or Ca at concentrations up to and beyond those innatural seawater. Reaction mixtures containing unin-duced cells were incubated for 3 h at 25 C and thosecontaining induced cells were incubated for 4.5 h at40 C. After incubation, the contents of the flasks wereacidified with 0.05 ml of 10 N H2SO4 for 10 min. Afteracidification, the liquid portion in each reaction vesselwas clarified by centrifugation and the supernatantwas assayed for the manganese released.

Determination of the effects of pre-exposing,uninduced whole cells to various cation solutions.For these experiments, uninduced cells were washedtwice in distilled water then resuspended in eitherdistilled water, 0.1 M NaCl, 0.059 M MgCl2.6HO, or0.059 M CaCl2. Measured portions of resultant sus-pensions were then transferred to appropriate reactionmixtures, giving a final cation concentration as speci-fied, and assayed for glucose-linked, ferricyanide-dependent MnO,-reductase activity. After incubationat 25 C for 3 h, the contents of each flask wereacidified, clarified by centrifugation, and assayed forreleased manganese.

Determination of the effect of temperature onMnO2-reductase activity. Uninduced whole cellswere washed and suspended in the medium in whichthey were tested (fivefold diluted seawater, 0.05 MMgCl2.6H2O or 0.1 M NaCl). Preparations of inducedcells were washed and suspended in fivefold dilutedseawater and tested in a reaction mixture made up inthat solution. Reaction mixtures were incubated attemperatures between 10 and 60 C. Those held attemperatures between 10 and 25 C were incubated inlow-temperature incubators. Those subjected to tem-peratures between 28 and 60 C were incubated inthermostatically controlled water baths. Uninducedcell preparations were incubated for 3 h and inducedcell preparations for 4.5 h. The temperature of allreaction mixtures was permitted to reach that of theroom before acidification. Inoculated mixtures wereincubated at the same time as blank controls for eachexperimental temperature point above 25 C since anincrease in nonenzymatic release of Mn2+ was ob-served in mixtures incubated at 30 C and above.Nonenzymatic release of Mn2+ in reaction mixturesincubated at temperatures below 25 C was consideredno greater than at 25 C. Therefore, the 25 C controlwas used as an estimate of the amount of Mn2+released in mixtures incubated at the lower tempera-tures. Since the number of experimental tempera-tures of incubation which could be tested in a singleexperiment was limited, results from several experi-

786 APPL. MICROBIOL.

VOL. 28, 1974 CATION AND TEMPERATURE EFFECTS ON MnO2-REDUCTASE

ments in which cells were incubated at differenttemperatures were pooled to study the effect oftemperature range on a given process.

RESULTSEffects of Na+, K+, Mg2+, and Ca2+ on

MnO,-reductase activity. The glucose-linked,ferricyanide-dependent, and the ferrocyanide-linked MnO2-reductase activities of uninducedBacillus 29 were stimulated by individual in-creases in concentration of each of the fourmajor cations of seawater, Na+, K+, Mg2+, andCa2+ (Fig. 1-3). Na+ and K+ exerted a qualita-

SEAWATER Di LUTIONliS 2/5 3/5 4/5

0.1 0.2 0.3 0.4 0.50.01 0.02 0.03 0.04 0.050.002 0.004 0.006 0.008 0.01

CATION CONCENTRATION(moles/ liter)

FIG. 1. Effects of various cation concentrationsand seawater dilutions on glucose-linked, ferricya-nide-dependent MnO,-reductase activity in unin-duced whole cells of Bacillus 29. For the tests withNa+, the cells were washed two times in distilledwater for one experiment and three times for a secondexperiment without a significant difference in results.For experiments with the other cations and seawaterdilutions, the cells were always washed three times indistilled water. The cells were added in a distilledwater suspension to each reaction mixture. The finalcation concentration of each reaction mixture was asspecified. Each reaction mixture contained 0.0004 MK+ from KFe(CN)., which probably accounts forminimal activity in the absence of other added salts.

tively similar effect on the glucose-linked activ-ity of whole cells (Fig. 1), cell extracts (Fig. 2),and the ferrocyanide-linked activity of cell ex-tracts (Fig. 3). Greatest stimulation was demon-strated in reaction mixtures containing concen-trations of the monovalent cations that wouldbe found in two- to fivefold diluted seawater.Mg2+ and Ca2+, on the other hand, exerteddissimilar effects on whole-cell and cell-extractactivities. The activity of whole cells was stimu-lated in a biphasic manner, being greatest in theassay mixtures containing the highest Mg2+ andc20=, , ,

UNINDUCED CELL EXTRACTS.FER0CYANIDE-LINKED ACTIVITY

00. 2.% 0-1 ~0.2 0-3 0-4 A apO 0.01 0.02 0.03 0.04 0.05 M)I0 0.002 0.004 0.006 0,006 0.01 K ,Cot+CATION CONCENTRATION (moles/liter)

FiG. 2. Effects of various cation concentrationsand seawater dilutions on glucose-linked, ferricya-nide-dependent MnO,-reductase activity in extractsof uninduced cells of Bacillus 29. The cells werewashed three times in distilled water and sonicated indistilled water. Each reaction mixture contained0.0004 M K+ from KFe(CN)., which probably ac-counts for minimal activity in the absence of otheradded salts.

SEAWATER DILUTIONf5 2/5 3/5 4/5

sv'O 0.1 0.2 0.3 0.4 0.5 Nas0 0.01 0.02 0.03 0.04 0.05 Mg2+0 0.002 0.004 0.006 0.008 0.01 KCa20

CATION CONCENTRATION (moles/liter)FIG. 3. Effects of various cation concentrations of

ferrocyanide-linked MnO2-reductase activity in ex-tracts of uninduced cells of Bacillus 29. Extracts wereprepared as for experiments in Fig. 2. Each reactionmixture contained 0.0022 M K+ fromK.Fe(CN)*.3H.0, which probably accounts for mini-mal activity in the absence of other salts.

787

GHIORSE AND EHRLICH

Ca2+ concentrations tested (Fig. 1, Table 1);whereas both activities in cell extracts were

stimulated strongly only up to approximately0.01 M Mg2+ and 0.002 M Ca2+ (Fig. 2 and 3).Further increases in the concentration of Ca2+and Mg2+ resulted in only minor stimulation.The molar concentrations of the four major

cations in seawater are approximately as fol-lows: Na+, 0.5 M; Mg2+, 0.05 M; Ca2+, 0.01 M;and K+, 0.01 M. At concentrations correspond-ing to those in fivefold diluted seawater, 0.01 MMg2+ or 0.002 M Ca2+ in the presence of 0.0022M K+ supported greater activity of uninducedwhole cells and cell extracts than 0.1 M Na+;and 0.1 M Na+ in the presence of 0.0022 M K+supported greater activity than 0.0022 M K+alone (Table 1). In the absence of other cations,0.01 M Mg2+ also supported greater activitythan 0.1 M Na+ alone. Indeed, the activity ofinduced cells was greatest in reaction mixturescontaining the highest Mg2+ concentrationtested (Table 2).Mixtures of Na+, Mg2+, and K+, or Na+,

Ca2+, and K+, or Na+, Mg2+, Ca2+, and K+,with each cation at a concentration equivalentto that in fivefold diluted seawater, supportedactivity in uninduced whole cells and extractsat about the same level as a mixture containing0.01 M Mg2+ plus 0.0022 M K+ exhibited. Theonly exception was the ferrocyanide-linked ac-

tivity of extracts in mixtures containing Na+,Mg2+, and K+, or Na+, Ca2+, and K+, whichwas depressed for unknown reasons (Table 1).Comparisons of these results showed that

fivefold diluted seawater was the most stimula-tory suspension medium for whole cells andextracts (Tables 1 and 2); the only exceptionswere solutions of 0.05 M Mg2+ and 0.025 M

TABLE 2. Effects of various cations in solutions and indiluted seawater on MnO2-reductase activity of whole

cells of induced Bacillus 29°

Sp actcReaction mixture" (nmol of Mn2+/h per

mg of protein)

Distilled water ..................... 7.00.1MNa+.......................... 10.30.01 M Mg2+........................ 15.90.05 M Mg2+........................ 18.80.25M Mg2+........................ 20.6One-tenth seawater .......... ...... 18.0One-fifth seawater ................. 23.2Six-sevenths seawater ........ ...... 2.2

a Cells were washed in distilled water before incu-bation.

b Chloride salts of elements yielding the appropri-ate cations were dissolved in distilled water.

cGlucose-linked activity of whole cells. Reactionmixtures were incubated at 40 C, which was found toprovide maximal activity (see Fig. 6).

TABLE 1. Effects of various cations in solutions and in diluted seawater on MnO2-reductase activity ofuninduced cells of Bacillus 29a

Sp act (nmol of Mn2+/h per mg

Reaction mixture" Whole cells Extracts

Glucosec Glucosec Fe(CN)- *d

0.0022 M K+ 42.0 5.8 + 0.3 8.3 + 0.20.1 M Na+ and 0.0022M K+ 98.0 6.8 + 0.1 9.4 + 0.40.01 M Mg2+ and 0.0022 M K+ 128.0 10.6 + 0.2 13.1 ± 0.50.05M Mg2+ and 0.0022 M K+ 168.00.25 M Mg2+ and 0.0022 M K+ 186.00.002 M Ca2+ and 0.0022 M K+ 135.0 10.0 + 0.60.05 M Ca2+ and 0.0022 M K+ 177.50.1 M Na+, 0.01 M Mg2+, and 0.0022 M K+ 146.5 11.2 ± 0.3 6.3 + 0.10.1 M Na+, 0.01 M Ca'+, and 0.0022M K+ 4.6 ± 0.60.1 M Na+, 0.01 M Mg2+, 0.002M Ca2+, 125.5 10.2 ± 0.5 11.4 ± 0.5and 0.0022 M K+

One-fifth seawater and 0.0022 M K+ 149.5 12.2 ± 0.5 12.0 ± 0.4Five-sevenths seawater and 0.0022M K+ 128.0

a Cells were washed in distilled water before incubation or sonication. Assays were performed at 25 C.Chloride salts of elements yielding the indicated cations were dissolved in distilled water. The K+

concentration was derived from 0.0004 M K+ from KFe(CN),, and from 0.0018 M K+ from KCI, or from0.0022 M K+ from K4Fe(CN), .3H20.

c Glucose-linked, ferricyanide-dependent activity.d Ferrocyanide-linked activity.

788 APPL. MICROBIOL.

VOL. 28, 1974 CATION AND TEMPERATURE EFFECTS ON MnO,-REDUCTASE

Mg2+ with uninduced whole cells. Lesser dilu-tions of seawater supported lower activities(Fig. 1 and 2, Tables 1 and 2), which may beattributed to the presence of high concentra-tions of Na+ and K+, both of which supportedless activity when used in concentrations foundin full-strength seawater (Fig. 1 and 2).Experiments designed to determine the ef-

fects of pre-exposure of uninduced whole cells tosolutions containing various cations (Table 3)showed that the ionic history of whole cellsaffected the MnO,-reductase activity subse-quently determined. Pre-exposure to mediacontaining 0.059 M Mg2+ or Ca2+ before assayproduced higher specific activity than pre-expo-sure to distilled water when 0.05 M Mg2+ orCa2+ was present in the reaction mixture (Table3). Furthermore, the higher specific activity wasnot diminished when the Mg2+ or Ca2+ concen-tration of the reaction mixture was lowered to0.008 M. Washing cells once with distilled waterafter Mg2+ pretreatment decreased the specificactivity to about 51% of the original valueobtained with 0.008 M Mg2+ in the reactionmixture. A second washing diminished the

TABLE 3. Effect of pre-exposure of whole cells ofBacillus 29 to various cations in solutions on

MnO,-reducing activity

Pre-exposurea Reaction Sp actcmixture"

Distilled water Distilled water 0Distilled water 0.05 M Mg2+ 249.00.059M Mg2+ 0.05 M Mg2+ 377.00.059 M Mg+ 0.008 M Mg2+ 399.00.05 M Mg'+; distilled 0.008 M Mg2+ 203.5

waterd0.059M Mg'+; distilled 0.008M Mg2+ 171.0

water'Distilled water 0.05 M Ca2+ 274.00.059M Ca2+ 0.05 M Ca2+ 302.00.059M Ca2+ 0.008M Ca2+ 301.0Distilled water 0.014M Na+ 59.0Distilled water 0.1 M Na+ 198.00.1MNa+ 0.014MNa+ 41.00.1MNa+ 0.1MNa+ 225.0

aWhole cells were washed twice in the solutionlisted.

"Includes 0.0004 M K+ from KFe(CN)..c Glucose-linked, ferricyanide-dependent MnO2-

reductase activity of uninduced whole cells. Specificactivity expressed as nanomoles of Mn2+ released perhour per milligram of cell protein. Reaction mixtureswere incubated at 25 C.

dCells washed with solutions containing Mg2+ weresubsequently washed once with distilled water.

'Cells washed twice with distilled water.

activity an additional 8% (Table 3). Theseresults indicate that Mg2+, and probably Ca2+,were rather tightly bound to cells. A similarretention of activity was not observed with 0.1M Na+ (Table 3) suggesting that this cation wasloosely bound, or not bound at all, by the cells.

Effect of temperature on MnO2-reductaseactivity. The influence of temperature on therate of glucose-linked, ferricyanide-dependentMnO,-reductase activity of uninduced wholecells was tested in fivefold diluted seawater, 0.1M NaCl, and 0.05 M MgCl2.6HO (Fig. 4). Inall three cases the optimal temperature for theactivity was in the range of 38 to 40 C. Althoughminimal and maximal temperatures were notdetermined in activity of uninduced whole cells,less than 20% of the maximal activity wasdetected at 15 and 50 C. It should be noted thatat 25 C (Fig. 4), the incubation temperaturemost commonly used in previous experiments,the specific activity of whole cells was onlyabout 45% of the maximal activity observed.

In extracts of uninduced cells, the optimaltemperature for the glucose-linked, ferricya-nide-dependent MnO2-reductase activity was30 C (Fig. 5A). Approximately 65% of thisactivity was detected at 15 C, whereas 40% wasdetected at 15 C and 40% at 50 C. The ferrocya-nide-linked MnO.-reductase activity of theseextracts exhibited a temperature optimum near40 C (Fig. 5B). The optimal temperature rangeof this activity was much broader than that ofthe glucose-linked, ferricyanide-dependent ac-tivities (Fig. 4 and 5A.)The optimal temperature of the glucose-

TEMPERATURE (0C)FIG. 4. Effect of incubation temperature on glu-

cose-linked, ferricyanide-dependent MnO.-reductaseactivity in uninduced whole cells of Bacillus 29.Specific activity was expressed as nanomoles of Mn2+released per hour per milligram of cell protein.

789

GHIORSE AND EHRLICH

linked activity in induced whole cells was alsofound to be near 40 C, with very little activitydetected at 15 or 50 C, (Fig. 6). In extracts thisactivity was greatest in mixtures incubated attemperatures near 25 C, with 70% of the maxi-mal activity detected in mixtures incubated at15 C and 60% at 10 C (Fig. 7). No activity wasdetected in induced cell extracts tested at 50 C.

DISCUSSIONThe effects of seawater cations which were

demonstrated in this work reflect the marineorigin of Bacillus 29. The finding that theMnO2-reductase system was most stimulatedby Na+ and K+ when these cations were presentat lower concentrations than in natural seawa-ter is not unlike the findings for enzymesderived from other marine bacteria (12-14).Although not specifically investigated, it is

appropriate to speculate about possible means

20

C)SU10

oOa:(

5A UNINDUCED CELLEXTRACTS

GLUCOSE-LINKED

SB UNINDUCED CELLEXTRACTS

FERROCYAN IDE-LINKED

10 20 30 40 50 60 10 20 30 40 50 60

TEMPERATURE (°C)FIG. 5. Effect of incubation temperature on glu-

cose-linked, ferricyanide-dependent and ferrocyanide-linked MnO2-reductase activities of extracts ofuninduced cells of Bacillus 29. (A) glucose-linked,ferricyanide-dependent activity; (B) ferrocyanide-linked activity. Specific activity was expressed as

nanomoles of Mn2+ released per hour per milligramofprotein.

>_20 FG F;1.NUCDWHOLE CELLS INDUCED CEL OHRC2%

0 010 20 30 40 50 60 10 20 30 40 50 60

TEMPERATURE (C)FIG. 6. Effect of incubation temperature on glu-

cose-linked MnO-reductase activity of uninducedwhole cells of Bacillus 29. Specific activity was

expressed as nanomoles of Mn2+ released per hourper milligram of cell protein.

FIG. 7. Effect of incubation temperature on glu-cose-linked MnO,-reductase activity of extracts ofinduced cells of Bacillus 29. Specific activity was

expressed as nanomoles of Mn2+ released per hour permilligram of protein.

by which the individual cations influence theMnO,-reducing activity of Bacillus 29. By anal-ogy with the effects of Na+ and K+ on othermarine bacteria and their enzymes (12, 13),these cations most likely affect the MnO,-reductase activity both by regulating the activ-ity of enzymes directly, and by affecting thepermeability of the cell membrane to com-pounds essential for activity. In support of thisanalogy are the results of experiments showingthat the Na+ concentration maxima were differ-ent for cell extracts (Fig. 2, about 0.1 M) andwhole cells (Fig. 1, about 0.25 M). In extracts,Na+ presumably acted by interacting with en-zymes of the system which were exposed to it. Inwhole cells, Na+ could also act in this way byaffecting either intracellular enzymes on the cellsurface such as the terminal MnO2-reductase,which must be exposed to the extracellularenvironment to react with MnO2. Alternatively,Na+ might affect the permeability of the cellmembrane to glucose or ferricyanide, thus in-directly regulating the activity. The fact thatboth the glucose- and ferrocyanide-linked sys-tems of cell extracts responded to various con-centrations of Na+ in qualitatively the sameway (Fig. 2 and 3) suggests that unless allenzymes in the electron transport system areequally sensitive to Na+, it is the terminal(ferrocyanide-linked) portion of the system,perhaps the terminal reductase itself, whichdetermines the response of the entire (glucose-linked) system to Na+. Thus, it is possible thatthe Na+ effect on the MnO2-reductase activityof whole cells is due to the combined responsesof the terminal enzyme of the system andmembrane permeability to the presence of thiscation.

Since K+ resembled Na+ in its effect on theMnO2-reductase system (i.e., Na+ and K+ con-centration maxima corresponded to those intwo- to fivefold diluted seawater [Fig. 1-3]), itseems likely that K+ may also affect both theenzymes of the system and membrane permea-bility. MacLeod (12, 13) has previously shownthat K+ affects permeability to organic sub-strates in some marine bacteria.Results showing that Ca2+ and Mg2+ elicited

a biphasic stimulation of whole-cell activity(Fig. 1), as well as stimulating the activity ofextracts (Fig. 2 and 3), suggest at least twopossible roles for these cations. One role may beactivation of the terminal reductase or otherenzymes in the cell membrane which affectglucose transport. Such enzymes, specificallyMg2+-activated adenosine triphosphatasewhich may function in active transport, have

790 APPL. MICROBIOL.

VOL. 28, 1974 CATION AND TEMPERATURE EFFECTS ON MnO,-REDUCTASE

been found in the isolated cell envelopes of twomarine pseudomonads and a Cytophaga (13).In addition, Mg2+ and Ca2+ may function toenhance binding of whole cells and/or the termi-nal MnO2-reductase to the mineral surface bythe formation of ionic bridges between negativecharges in the cell-wall matrix, or on the en-zyme particles, (e.g., carboxyl and phosphategroups) and like charges on the surface of theMnO2 particles. Such enhanced binding wouldbring the terminal reductase into closer prox-imity to its solid substrate, thus elevating theactivity.The binding-enhancement hypothesis is sup-

ported indirectly by pre-exposure experimentswhich indicated that Mg2+ and Ca2+ were muchmore tightly bound to the cells than Na+. Theseresults parallel those of Cutinelli and Galdiero(1) who showed that the cell walls of Staphylo-coccus aureus behaved like weak cation-exchange resins, binding both monovalent anddivalent cations by ionic bonds. Further, stud-ies by Galdiero et al. (10) showed that thedivalent cations were strongly bound by the cellwalls, presumably due to electrostatic interac-tions with negative charges in the wall matrix.The monovalent cations were less stronglybound, appearing to form a mobile monolayeraround the cell wall surface.Thus, Na+, K+, Ca2+, and Mg24 may exert

multiple effects on the MnO2-reducing activityof whole cells by influencing membranepermeability, active transport, and binding ofenzymes or the cells to MnO, particles, as wellas by directly regulating the activity of enzymesof the MnO2-reductase system itself.Experiments testing the influence of various

temperatures on the rate of MnO2-reductaseactivity (Fig. 4-7) indicated that the tempera-ture optimum of the MnO2-reducing system haschanged over the past five years. Previousexperiments with induced and uninduced wholecells showed that the optimal temperatures oftheir respective MnO2-reductase activities were25 and 30 C (unpublished data). Trimble andEhrlich (16) found that the induced system incell extracts was most active at 18 C. Thepresent optimum for both induced and unin-duced whole cells is 40 C (Fig. 5 and 6), and forinduced cell extracts is 25 C (Fig. 7). Thesefindings support earlier conclusions (7; Trim-ble, Ph.D. thesis, Rensselaer Polytechnic Insti-tute, Troy, N.Y. 1969) that the temperaturecharacteristics of Bacillus 29 have changedsince its isolation more than 10 years ago. Thechanges may be attributable to the selectiveeffects of cultivation of populations of this

organism at 25 C, which is higher than thetemperature of its original marine habitat (5 to10 C). Presumably, strains have developed fromthe original culture under the selective pressureof higher growth temperature which haveMnO,-reducing enzymes with higher tempera-ture optima than the parent culture.The current experiments also indicate varia-

tion in the relative heat stabilities of the termi-nal and inducible portions of the MnO2-reduc-tase system. The terminal (ferrocyanide-linked)portion of the system was less inhibited bytemperatures of 50 and 60 C than the entireglucose-linked, ferricyanide-dependent systemin cell extracts (Fig. 5A and B). This suggeststhat the terminal enzyme component is rela-tively more heat stable than other enzymes ofthe system. Heat-inactivation experiments,showing that the ferrocyanide-linked activity insoluble enzyme fraction of uninduced Bacillus29 retained 60% of its original activity after 20min in boiling water, whereas the glucose-linkedactivity of uninduced whole cells was com-pletely inactive after 5 min of boiling (Ghiorse,Ph.D. thesis, Rensselaer Polytechnic Institute,Troy, N.Y. 1972), are consistent with the forego-ing conclusion. The findings that enzymes ininduced whole cells and extracts were inactiveat 50 C (Fig. 6 and 7), whereas those in unin-duced cells and extracts still had measurableresidual activity at that temperature (Fig. 4 and5), indicate that the inducible portion of thesystem is more heat-sensitive than other partsof the system. This heat sensitivity could ex-plain why induced cell extracts had a lowertemperature optimum than uninduced extracts.The temperature characteristics of the MnO2-

reductase system of Bacillus 29 are unusual.MnO,-reducing isolates from the Pacific Oceanstudied in this laboratory exhibit stable temper-ature optima more in keeping with their originalhabitat (Ehrlich, unpublished data).

ACKNOWLEDGMENTSW. C. Ghiorse was supported by a National Aeronautics

and Space Administration traineeship during this work.We wish to thank Rosemary G. Ghiorse and Reggie

Malinowski for their help in preparing the manuscript.

LITERATURE CITED1. Cutinelli, C., and F. Galdiero. 1967. Ion-binding proper-

ties of the cell wall of Staphylococcus aureus. J.Bacteriol. 93:2022-2023.

2. Ehrlich, H. L. 1963. Bacteriology of manganese nodules.I. Bacterial action on manganese in nodule enrich-ments. Appl. Microbiol. 11:15-19.

3. Ehrlich, H. L. 1964. Microbial transformation of min-erals, p. 43-60. In H. Heukelekian and N. C. Dondero(ed.), Principles and applications in aquatic microbiol-

791

792 GHIORSE AND EHRLICH

ogy. John Wiley and Sons, Inc., N.Y.4. Ehrlich, H. L. 1966. Reactions with manganese by

bacteria from marine ferromanganese nodules. Dev.Ind. Microbiol. 7:279-286.

5. Ehrlich, H. L. 1968. Bacteriology of manganese nodules.II. Manganese oxidation by cell-free extract from amanganese nodule bacterium. Appl. Microbiol.16:197-202.

6. Ehrlich, H. L. 1970. The microbiology of manganesenodules. U.S. Clearinghouse, Fed. Sci. Tech. Inform.no. 716,508.

7. Ehrlich, H. L. 1971. Bacteriology of manganese nodules.V. Effect of hydrostatic pressure on bacterial oxidationof Mn" and reduction of MnO.. Appl. Microbiol.21:306-310.

8. Ehrlich, H. L. 1974. Response of some activities offerromanganese nodule bacteria to hydrostatic pres-sure. In R. R. Colwell and R. Y. Morita (ed.), Effect ofthe ocean environment on microbial activities. Univer-sity Park Press, Baltimore, Md.

9. Ehrlich, H. L., S. H. Yang, and J. D. Mainwaring, Jr.1973. Bacteriology of manganese nodules. VI. Fate ofcopper, nickel, cobalt, and iron during bacterial andchemical reduction of the manganese (IV). Z. Allg.

APPL. MICROBIOL.

Mikrobiol. 13:39-48.10. Galdiero, F., M. Lembo, and M. A. Tufano. 1968. Affinity

of various cations for Staphylococcus aureus cell-wall.Experientia 24:34-36.

11. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951. Protein measurement with the Folinphenol reagent. J. Biol. Chem. 193:265-275.

12. MacLeod, R. A. 1965. The question of the existence ofspecific marine bacteria. Bacteriol. Rev. 29:9-23.

13. MacLeod, R. A. 1968. On the role of inorganic ions in thephysiology of marine bacteria. Advan. Microbiol. Sea1:95-126.

14. MacLeod, R. A., and A. Hori. 1960. Nutrition andmetabolism of marine bacteria. VIII. Tricarboxylicacid cycle-enzymes in a marine bacterium and theirresponse to added salts. J. Bacteriol. 80:464-471.

15. Trimble, R. B., and H. L. Ehrlich. 1968. Bacteriology ofmanganese nodules. III. Reduction of MnO, by twostrains of nodule bacteria. Appl. Microbiol. 16:695-702.

16. Trimble, R. B., and H. L. Ehrlich. 1970. Bacteriology ofmanganese nodules. IV. Induction of an MnO,-reduc-tase system in a marine Bacillus. Appl. Microbiol.19:966-972.